Let f : N → N be a strictly increasing function such that f ( 2 ) = 2 and f ( a b ) = f ( a ) ⋅ f ( b ) for g cd ( a , b ) = 1 .
Evaluate the sum of all possible values of f ( 1 7 ) .
This section requires Javascript.
You are seeing this because something didn't load right. We suggest you, (a) try
refreshing the page, (b) enabling javascript if it is disabled on your browser and,
finally, (c)
loading the
non-javascript version of this page
. We're sorry about the hassle.
Another similar reasoning which leads to f ( 3 ) can be as follows. Suppose that f ( a ) = a for some a . We see that f ( x ) = x ∀ x ≥ a , otherwise, if f ( b ) = b for some b > a , we can use the strictly increasing condition to arrive at f ( a ) = a . So, let a be the smallest number such that f ( a ) = a . By hypothesis, f ( k ) = k for all k < a . But if a > 3 , we can always find two numbers c and d , both less than a , such that c d > a , g cd ( c , d ) = 1 and f ( c d ) = f ( c ) ⋅ f ( d ) = c d , contradicting the assumption that f ( x ) = x ∀ x ≥ a .
Log in to reply
Almost. If a is the least integer such that f ( a ) = a , and if a > 3 then a − 1 , a − 2 < a and so f ( ( a − 1 ) ( a − 2 ) ) = f ( a − 1 ) f ( a − 2 ) = ( a − 1 ) ( a − 2 ) , but ( a − 1 ) ( a − 2 ) ≥ a . This contradiction shows that a cannot be greater than 3 . We still have to show that a cannot be 3 , and hence that a does not exist.
Log in to reply
Yes, I was referring to reduction till f ( 3 ) . These approaches remind me of finite descent.
Note that, f ( 2 m + 1 ) ≤ f ( 2 ( 2 m − 1 + 1 ) ) = 2 f ( 2 m − 1 + 1 ) ≤ . . . = 2 m f ( 2 ) = 2 m + 1
Also, for any x , we can find m such that 2 m ≤ x ≤ 2 m + 1 . Thus 2 x ≥ 2 m + 1 and f ( x ) ≤ f ( 2 m + 1 + 1 ) ≤ 2 m + 2
⟹ x f ( x ) ≤ 4
Now, assume that f ( x ) = x . So, there is some k > 2 such that f ( k ) = k . Since f is strictly increasing, it follows that f ( k ) > k and f ( x ) > x ∀ x ≥ k
Denote the primes greater than k as p 1 , p 2 , p 3 , … . Define c n = p 1 ⋅ p 2 ⋅ … ⋅ p n
Now, f ( p i ) ≥ p i + 1 for any p i since f ( x ) > x ∀ x ≥ k . So, we have,
f ( c n ) = f ( p 1 ) . . . f ( p n ) ≥ ( p 1 + 1 ) . . . ( p n + 1 )
⟹ c n f ( c n ) ≥ ( 1 + p 1 1 ) ⋅ ( 1 + p 2 1 ) ⋅ … ⋅ ( 1 + p n 1 ) ≥ ( p 1 1 + p 2 1 + … + p n 1 ) + …
Since the sum of the reciprocals of the prime numbers diverges and only finitely many primes are missing, it follows that the product on the RHS diverges as n → ∞ . Thus the sequence c n f ( c n ) → ∞ , which is impossible for the bounded sequence c n f ( c n ) .
∴ f ( x ) = x ∀ x ∈ N
f ( 1 5 ) < f ( 1 8 )
f ( 3 ) f ( 5 ) < 2 f ( 9 ) < 2 f ( 1 0 ) = 4 f ( 5 )
2 < f ( 3 ) < 4
f ( 3 ) = 3
f ( 6 ) = 6 , so we must have f ( n ) = n for n = 4 , 5
f ( 3 0 ) = 3 0 , so f ( 1 7 ) = 1 7
Problem Loading...
Note Loading...
Set Loading...
We shall show below that f ( 3 ) = 3 . Assuming that, we can show that f ( n ) = n for al n ≥ 1 . Now f ( 6 ) = f ( 2 ) f ( 3 ) = 6 and so 3 < f ( 4 ) < f ( 5 ) < 6 , so that f ( 4 ) = 4 , f ( 5 ) = 5 . Certainly, then, f ( j ) = j for all 1 ≤ j ≤ 5 .
Suppose inductively that f ( j ) = j for all 1 ≤ j ≤ 4 n + 1 . Then 4 n + 6 = 2 ( 2 n + 3 ) and 2 n + 3 ≤ 4 n + 1 , so that f ( 4 n + 6 ) = f ( 2 ) f ( 2 n + 3 ) = 4 n + 6 , and hence 4 n + 1 = f ( 4 n + 1 ) < f ( 4 n + 2 ) < f ( 4 n + 3 ) < f ( 4 n + 4 ) < f ( 4 n + 5 ) < 4 n + 6 so that f ( 4 m + 2 ) = 4 n + 2 , f ( 4 n + 3 ) = 4 n + 3 , f ( 4 n + 4 ) = 4 n + 4 and f ( 4 n + 5 ) = 4 n + 5 . Thus we have shown that f ( j ) = j for all 1 ≤ j ≤ 4 n + 5 . Thus we deduce by induction that f ( j ) = j for all j ≥ 1 . Thus the only possible value of f ( 1 7 ) is 1 7 .
It remains to show that f ( 3 ) = 3 .
Suppose first that f ( 3 ) = 4 . Then f ( 4 ) ≥ 5 , f ( 5 ) ≥ 6 , and hence f ( 6 0 ) = f ( 3 ) f ( 4 ) f ( 5 ) ≥ 1 2 0 . On the other hand, f ( 6 ) = f ( 2 ) f ( 3 ) = 8 , so f ( 5 ) ≤ 7 , so f ( 1 0 ) = f ( 2 ) f ( 5 ) ≤ 1 4 , so f ( 9 ) ≤ 1 3 , so f ( 1 8 ) = f ( 2 ) f ( 9 ) ≤ 2 6 , so f ( 1 7 ) ≤ 2 5 , so f ( 3 4 ) = f ( 2 ) f ( 2 7 ) ≤ 5 0 , so f ( 3 3 ) ≤ 4 9 , so f ( 6 6 ) = f ( 2 ) f ( 3 3 ) ≤ 9 8 , and hence f ( 6 0 ) ≤ 9 2 . This contradiction implies that f ( 3 ) = 4 .
Suppose that f ( 3 ) = n ≥ 5 . Then f ( 4 ) ≥ n + 1 , f ( 5 ) ≥ n + 2 , and hence f ( 6 0 ) ≥ n ( n + 1 ) ( n + 2 ) ≥ 4 2 n . On the other hand f ( 6 ) = 2 n , so f ( 5 ) ≤ 2 n − 1 , so f ( 1 0 ) ≤ 4 n − 2 , so f ( 9 ) ≤ 4 n − 3 , so f ( 1 8 ) ≤ 8 n − 6 , so f ( 1 7 ) ≤ 8 n − 7 , so f ( 3 4 ) ≤ 1 6 n − 1 4 , so f ( 3 3 ) ≤ 1 6 n − 1 5 , so f ( 6 6 ) ≤ 3 2 n − 3 0 , so f ( 6 0 ) ≤ 3 2 n − 3 6 ≤ 3 2 n . This contradiction shows that this option is impossible as well.
Thus we must have f ( 3 ) = 3 , and so the proof is complete.